ProtoAnomalyNCD: Prototype Learning for Multi-class Novel Anomaly Discovery in Industrial Scenarios

TL;DR

ProtoAnomalyNCD introduces a prototype learning framework that effectively discovers and classifies multiple unseen anomaly types in industrial scenarios by leveraging localized priors and attention mechanisms, outperforming existing methods.

Contribution

The paper proposes a novel prototype-learning-based framework that integrates localized priors and attention modules for multi-class anomaly discovery and classification in industrial settings.

Findings

Outperforms state-of-the-art on MVTec AD, MTD, and Real-IAD datasets.

Effectively discovers and clusters unseen anomaly classes.

Enables multi-type anomaly classification and outlier detection.

Abstract

Existing industrial anomaly detection methods mainly determine whether an anomaly is present. However, real-world applications also require discovering and classifying multiple anomaly types. Since industrial anomalies are semantically subtle and current methods do not sufficiently exploit image priors, direct clustering approaches often perform poorly. To address these challenges, we propose ProtoAnomalyNCD, a prototype-learning-based framework for discovering unseen anomaly classes of multiple types that can be integrated with various anomaly detection methods. First, to suppress background clutter, we leverage Grounded SAM with text prompts to localize object regions as priors for the anomaly classification network. Next, because anomalies usually appear as subtle and fine-grained patterns on the product, we introduce an Anomaly-Map-Guided Attention block. Within this block, we design a Region Guidance Factor that helps the attention module distinguish among background, object regions, and anomalous regions. By using both localized product regions and anomaly maps as priors, the module enhances anomalous features while suppressing background noise and preserving normal features for contrastive learning. Finally, under a unified prototype-learning framework, ProtoAnomalyNCD discovers and clusters unseen anomaly classes while simultaneously enabling multi-type anomaly classification. We further extend our method to detect unseen outliers, achieving task-level unification. Our method outperforms state-of-the-art approaches on the MVTec AD, MTD, and Real-IAD datasets.